[go: up one dir, main page]
More Web Proxy on the site http://driver.im/

US8488061B2 - Deriving video signatures that are insensitive to picture modification and frame-rate conversion - Google Patents

Deriving video signatures that are insensitive to picture modification and frame-rate conversion Download PDF

Info

Publication number
US8488061B2
US8488061B2 US12/600,466 US60046608A US8488061B2 US 8488061 B2 US8488061 B2 US 8488061B2 US 60046608 A US60046608 A US 60046608A US 8488061 B2 US8488061 B2 US 8488061B2
Authority
US
United States
Prior art keywords
video signal
series
resolution
video
pictures
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Fee Related, expires
Application number
US12/600,466
Other languages
English (en)
Other versions
US20100238350A1 (en
Inventor
Regunathan Radhakrishnan
Claus Bauer
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Dolby Laboratories Licensing Corp
Original Assignee
Dolby Laboratories Licensing Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Dolby Laboratories Licensing Corp filed Critical Dolby Laboratories Licensing Corp
Priority to US12/600,466 priority Critical patent/US8488061B2/en
Publication of US20100238350A1 publication Critical patent/US20100238350A1/en
Application granted granted Critical
Publication of US8488061B2 publication Critical patent/US8488061B2/en
Expired - Fee Related legal-status Critical Current
Adjusted expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11B—INFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER
    • G11B27/00—Editing; Indexing; Addressing; Timing or synchronising; Monitoring; Measuring tape travel
    • G11B27/10—Indexing; Addressing; Timing or synchronising; Measuring tape travel
    • G11B27/19—Indexing; Addressing; Timing or synchronising; Measuring tape travel by using information detectable on the record carrier
    • G11B27/28—Indexing; Addressing; Timing or synchronising; Measuring tape travel by using information detectable on the record carrier by using information signals recorded by the same method as the main recording
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
    • G06F16/78—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/783—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
    • G06F16/7847—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using low-level visual features of the video content
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T1/00—General purpose image data processing
    • G06T1/0021—Image watermarking
    • G06T1/0028—Adaptive watermarking, e.g. Human Visual System [HVS]-based watermarking
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T1/00—General purpose image data processing
    • G06T1/0021—Image watermarking
    • G06T1/005—Robust watermarking, e.g. average attack or collusion attack resistant
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/40—Scenes; Scene-specific elements in video content
    • G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/46—Embedding additional information in the video signal during the compression process
    • H04N19/467—Embedding additional information in the video signal during the compression process characterised by the embedded information being invisible, e.g. watermarking
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2201/00—General purpose image data processing
    • G06T2201/005—Image watermarking
    • G06T2201/0051—Embedding of the watermark in the spatial domain
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2201/00—General purpose image data processing
    • G06T2201/005—Image watermarking
    • G06T2201/0061—Embedding of the watermark in each block of the image, e.g. segmented watermarking

Definitions

  • the present invention pertains generally to the processing of video signals and pertains more specifically to processes that extract features from video signals to identify the signals.
  • video signals and “video content” refer to signals and content that represent images intended for visual perception.
  • Examples of unintentional modifications to signal content include the insertion or addition of noise to signals in transmission channels and on storage media.
  • intentional modifications to video signals include luminance and color modifications such as contrast/brightness adjustments, gamma correction, luminance histogram equalization, color saturation adjustments and color correction for white balancing, include geometric modifications such as image cropping and resizing, image rotation and flipping, stretching, speck removal, blurring, sharpening and edge enhancement, and include coding techniques such as lossy compression and frame rate conversion.
  • FIGS. 1 and 2 are schematic block diagrams of a video signature generator that may be used to obtain a reliable identification of a video signal.
  • FIG. 3 is a schematic block diagram of a process performed in one implementation of an image pre-processor.
  • FIG. 4 is a schematic block diagram of a lower-resolution image obtained by a spatial-domain processor.
  • FIG. 5 is a schematic block diagram of video frames arranged in segments.
  • FIG. 6 is a schematic block diagram of a video signature generator processing segments of video content to generate a set of video signatures.
  • FIG. 7 is a schematic block diagram of a system that manages a signature data base for detection of copies of video content.
  • FIG. 8 is a schematic block diagram of a device that may be used to implement various aspects of the present invention.
  • Various aspects of the present invention may be used advantageously in a system for identifying video content by analyzing segments of that content and generating a signature for each segment.
  • the signatures generated for the segments in an interval of a signal form a signature set, which can be used as a reliable identification of the content in that interval.
  • the following disclosure first describes processes that may be used to generate a signature for a single segment and then describes the generation and use of signature sets.
  • FIG. 1 is a schematic block diagram of a video signature generator 100 that analyzes the video content in a signal segment 3 to generate a video signature 193 that identifies or represents that content.
  • the segment 3 includes a series of video frames 3 a to 3 d .
  • an audio signature that represents the audio content may be obtained by processing the audio content in a variety of ways including those disclosed in U.S. provisional patent application No. 60/872,090 entitled “Extracting Features of Video and Audio Signal Content to Provide a Reliable Identification of the Signals” by Regunathan Radhakrishnan, et al., filed on Nov. 30, 2006, which is incorporated herein by reference in its entirety.
  • an image pre-processor 110 obtains a series of format-independent images for the pictures conveyed in the frames 3 a , 3 b , 3 c , 3 d
  • a spatial-domain processor 130 down-samples the format-independent images to generate a series of lower-resolution representations of the format-independent images
  • a temporal-domain processor 150 generates values that represent a composite of the series of lower-resolution representations
  • a video signature processor 170 applies a hash function to the composite values to generate the video signature 193 that represents and identifies the content of the segment 3 .
  • the processing that is performed by the processors 110 , 130 , 150 and 170 may be implemented in a variety of ways. Preferred implementations of these processes are described below.
  • each video frame 3 a , 3 b , 3 c , 3 d in the segment 3 conveys a picture that is represented by an array of pixels D.
  • the image pre-processor 110 derives a format-independent image of the picture for each frame.
  • the format-independent image is represented by an array of pixels F.
  • the derivation of the format-independent image may be done in a variety of ways. A few examples are described below.
  • the video signature generator 100 generates signatures for television video signals that convey video content in a variety of formats including progressive-scan and interlaced-scan with the standard-definition (SD) resolution of 480 ⁇ 640 pixels and the high-definition (HD) resolution of 1080 ⁇ 1920 pixels.
  • SD standard-definition
  • HD high-definition
  • the image pre-processor 110 converts the picture in each frame into a format-independent image that has a format common to all signal formats of interest.
  • the pixels F in the format-independent images are obtained by down-sampling the pixels D in the frame to reduce sensitivity to modifications that can occur when frames of video are converted between different formats.
  • the resolution of the format-independent image is chosen to have a resolution of 120 ⁇ 160 pixels, which is a convenient choice for television signals conveying images in HD and SD resolutions for both progressive-scan interlaced-scan formats.
  • the image pre-processor 110 converts SD-format video content into format-independent images by down-sampling the pixels in each frame picture by a factor of four.
  • the image pre-processor 110 converts HD-format video content into format-independent images by cropping each frame picture to remove 240 pixels from the left-hand edge and 240 pixels from right-hand edge to obtain an interim image with a resolution of 1080 ⁇ 1440 pixels and down-sampling the pixels in the interim image by a factor of nine.
  • a video signal conveys content in an interlaced-scan format in which frames of video are arranged in two fields
  • the signal may be converted into a progressive-scan format before obtaining the format-independent image.
  • greater independence from the choice of scan format can be achieved by obtaining the format-independent image from only one of the fields in an interlaced-scan frame.
  • the format-independent image can be obtained from only the first field in each frame or from only the second field in each frame. Video content in the other field can be ignored. This process avoids the need to convert to a progressive-scan format before obtaining the format-independent image.
  • the resultant image is essentially independent of the frame picture format so that the subsequent signature generation process is insensitive to different formats and to modifications that occur from conversions between formats.
  • This approach increases the likelihood that a video signature generated from a series of format-independent images will correctly identify the video content in a series of frame pictures even if those pictures have been subjected to format conversion.
  • the format-independent image excludes picture areas that are likely to be affected by intentional modifications.
  • this may be achieved by cropping to exclude corners and edges of the image where logos or other graphical objects may be inserted into the video content.
  • FIG. 3 provides a schematic illustration of the results obtained by a process 112 performed by the image pre-processor 110 that includes the cropping and down-sampling operations described above.
  • the picture in the frame 3 a within the segment 3 is cropped to extract the pixels D in a central portion of the picture.
  • the pixels D in this central portion are down-sampled to obtain the pixels F in the format-independent image 5 a .
  • a format-independent image 5 a , 5 b , 5 c , 5 d in a series of images 5 is obtained for each frame 3 a , 3 b , 3 c , 3 d in the segment 3 .
  • IP[ ] the image pre-processor operations applied to the picture in frame m;
  • M the number of frames in the segment.
  • the cropping operation that resizes a picture for format conversion may be combined with or performed separately from the cropping operation that excludes areas of a picture that may be affected by intentional modification such as the insertion of logos.
  • the cropping operations may be performed before or after the down-sampling operations.
  • the format-independent image may be obtained by cropping video content and subsequently down sampling the cropped images, it can be obtained by down sampling the video content and subsequently cropping the down-sampled images, and it can be obtained by a down-sampling operation performed between the two cropping operations mentioned above.
  • each video frame conveys a color image comprising pixels represented by red, green and blue (RGB) values
  • a separate format-independent image may be obtained for each of the red, green, and blue values in each frame.
  • one format-independent image is obtained for each frame from the luminance or brightness of pixels that is derived from the red, green, and blue values in the frame.
  • the format-independent image may be obtained from the intensities of the individual pixels in that frame.
  • the spatial-domain processor 130 obtains a down-sampled lower-resolution representation of the format-independent images by grouping the pixels F in each of the format-independent images into regions that are GX pixels wide and GY pixels high.
  • a lower-resolution image with picture elements E is derived from the intensities of the pixels F in a respective format-independent image by calculating the average intensity of the pixels in each region.
  • Each lower-resolution image has a resolution of K ⁇ L elements. This is illustrated schematically in FIG. 4 .
  • the picture elements E may be obtained by performing a process that implements the following expression:
  • E m (k,l) a picture element in the lower-resolution image for frame m;
  • GX the width of pixel groups expressed in numbers of pixels F
  • GY the height of pixel groups expressed in numbers of pixels F
  • K the horizontal resolution of the lower-resolution image
  • F m (i,j) a pixel in the format-independent image for frame m.
  • RH and RV are the horizontal and vertical resolutions of the format-independent image, respectively.
  • the grouping performed by the spatial-domain processor 130 can be combined with or performed prior to processing performed by the image pre-processor 110 .
  • the generated video signature is less sensitive to processes that change details of video signal content but preserve average intensity.
  • values that represent a composite of the series of lower-resolution images are obtained from the temporal averages and variances of respective picture elements E.
  • the temporal average Z(k,l) of each respective picture element E(k,l) may be calculated from the following expression:
  • the video content of selected frames within the segment 3 may be given greater importance by calculating the temporal averages from a weighted sum of the picture elements as shown in the following expression:
  • w m the weighting factor for picture elements in the lower-resolution image derived from the video content of frame in.
  • time-domain process represented by expression 3a or 3b may be performed prior to the spatial-domain process represented by expression 2.
  • the value Z(k,l) represents an average intensity for each picture element E(k,l) over both time and space; therefore, these average values do not convey much information about any motion that may be represented by the video content of the segment 3 .
  • a representation of motion may be obtained by calculating the variance of each picture element E(k,l).
  • the variance V(k,l) of each respective picture element E(k,l) may be calculated from the following expression:
  • the variance V(k,l) of each respective picture element E(k,l) may be calculated from the following expression:
  • the values that represent a composite of the series of lower-resolution images are the values of elements in two rank matrices Z, and V, that are derived from the temporal average and variance arrays Z and V, respectively.
  • the value of each element in the rank matrices represents the rank order of its respective element in the associated arrays. For example, if the element Z(2,3) is the fourth largest element in the average value array Z, the value of the corresponding element Z(2,3) in the rank matrix Z r is equal to 4.
  • the values that represent a composite of the series of lower-resolution images are the values of the elements in the temporal average and variance arrays Z and V.
  • the video signature processor 170 applies a hash function to K ⁇ L arrays of the composite values QZ and QV to generate two sets of hash bits. A combination of these two sets of hash bits constitute the video signature that identifies the content of the segment 3 .
  • the hash function is relatively insensitive to changes in the composite values and more sensitive to changes in any hash key that may be used.
  • a preferred hash function for this application provides an output that undergoes only small changes for small changes in the input composite values. This allows the generated video signature to change only slightly with small changes to video content.
  • One suitable hash function uses a set of N z base matrices to generate a set of N z hash bits for the QZ composite values, and uses a set of N V base matrices to generate a set of N V hash bits for the QV composite values.
  • Each of the base matrices is a K ⁇ L array of elements. These elements represent a set of vectors that preferably are orthogonal or nearly orthogonal to one another. In the implementation described below, the elements of the base matrices are generated by a random-number generator under the assumption that these elements represent a set of vectors that are nearly orthogonal to one another.
  • the generator RNG generates random or pseudo-random values that are uniformly distributed in the range [0,1].
  • the initial state of the generator may be initialized by a hash key, which allows the hash function and the generated video signature to be cryptographically more secure.
  • One set of hash bits BZ n is obtained by first projecting the composite values QZ onto each of the N z base matrices, which may be expressed as:
  • HZ n the projection of the composite values QZ onto the base matrix PZ n .
  • the set of hash bits BZ n is then obtained by comparing each projection to the median value of all projections and setting the hash bit to a first value if the projection is equal to or exceeds the threshold and setting the hash bit to a second value if the projection is less than the threshold.
  • This process may be expressed as: where
  • H Z the median value of all projections HZ n .
  • BV n sgn( HV n ⁇ H v ) (14)
  • HV n the projection of the composite values QV onto the base matrix PV n ;
  • H v the median value of all projections HV n .
  • the video signature is obtained from a concatenation of the two sets of hash bits, which forms a value that has a total bit length equal to N Z +N V .
  • the values for N Z and N V may be set to provide the desired total bit length as well as weight the relative contribution of the composite values QZ and QV to the final video signature.
  • N Z and N V are both set equal to eighteen.
  • a signature generated by the video signature generator 100 represents the video content of the segment from which the signature was generated.
  • a reliable identification of the video content in an interval of a signal much longer than a segment can be obtained by generating a set of signatures for the segments included in that interval.
  • the diagram shown in FIG. 5 is a schematic illustration of an interval of a signal that includes several segments of video frames. Five segments are shown.
  • the first segment 3 of the signal includes video frames 3 a to 3 d .
  • Each subsequent segment 4 , 5 , 6 , 7 includes video frames 4 a to 4 d , 5 a to 5 d , 6 a to 6 d and 7 a to 7 d , respectively.
  • a set of signatures can be generated for these segments by using the video signal generator 100 to process the contents of the video frames in each segment as described above.
  • Each segment contains an integral number of video frames.
  • the series of frames in each segment conveys video content for an interval of time that is equal to a nominal length L or within one frame period of the nominal length L.
  • the term “frame period” refers to the duration of the video content conveyed by one frame.
  • the nominal start times t# for successive segments are separated from one another by an offset ⁇ T.
  • This offset may be set equal to the frame period of the lowest frame rate of signals to be processed by the video signature generator 100 . For example, if the lowest rate to be processed is twelve frames per second, the offset ⁇ T may be set equal to 1/12 sec. or about 83.3 msec.
  • the nominal length L may be chosen to balance competing interests of decreasing the sensitivity of the subsequently-generated video signature to content modifications such as frame-rate conversion and increasing the temporal resolution of the representation provided by the video signature. Empirical studies have shown that a nominal segment length L that corresponds to about two seconds of video content provides good results for many applications.
  • the offset between the actual start times of successive segments can vary as shown in the figure by the different offset amounts ⁇ 1 and ⁇ 2 . If desired, the length of the offset between actual start times may kept within one frame period of the nominal offset ⁇ T.
  • FIG. 6 is a schematic block diagram showing a set of video signatures 193 to 197 that are generated from the video content of segments 3 to 7 , respectively.
  • the video signature generator 100 obtains the video content of the segment 3 starting at the nominal start time t 1 and processes this video content to generate the video signature 193 .
  • the video signature generator 100 then obtains the video content of the segment 4 starting at the nominal start time t 2 and processes this video content to generate the video signature 194 .
  • the generator continues by processing the video content in segments 5 , 6 and 7 , which begin at nominal start times t 3 , t 4 and t 5 , respectively, to generate the video signatures 195 , 196 and 197 .
  • Signatures may be generated for essentially any number of segments that may be desired.
  • the nominal start times do not need to correspond to any particular time data that may accompany the video content.
  • the alignment between the nominal start times and the video content is arbitrary.
  • the nominal start times are expressed as relative offsets from the beginning of a signal to be processed.
  • Each segment begins with the video frame conveying video content having a start time that is closest to its respective nominal start time.
  • each segment could begin with the video frame that spans the nominal start time for that segment.
  • any alignment between beginning frame and nominal start time may be used.
  • the signature sets generated from segments of video content can be used to identify the content even when that content has been modified by a variety of processes including those mentioned above.
  • the ability to determine reliably whether specified video content is a copy of a reference content, even when modified, can be used in a variety of ways including the following:
  • FIG. 7 is a schematic block diagram of a system that may be used to implement a variety of applications such as those mentioned in the preceding list.
  • the video signature generator 100 generates reference video signature sets from reference streams of video content received from the path 31 .
  • the generated reference video signature sets are stored in the signature data base 180 .
  • the reference signature sets may be stored with other information that may facilitate implementation of the application.
  • the reference signature sets may be stored with the underlying content itself or with information about the content such as the content owner, content licensing terms, title of the content or a textual description of the content.
  • Each reference signature set has a data base search key. This key may be derived in any manner that may be desired. Preferably, the key is based on or derived from the signatures in the associated reference signature set.
  • Any specified video content may be checked against reference content represented by one or more signature sets stored in the signature data base.
  • the content to be checked is referred to herein as the test content.
  • the identity of the test video content may be checked by having the video signature generator 101 generate one or more test video signature sets from the test video content received from the path 33 and passing the test video signature sets to the video search engine 185 .
  • the video search engine 185 attempts to find reference video signature sets in the signature data base 180 that are exact or close matches to the test video signature sets.
  • the video search engine 185 receives one or more test signature sets from the video signature generator 101 .
  • Each test signature set includes an ordered series of test signatures S TEST in the order in which they were generated from the test content.
  • the video search engine 185 receives reference signature sets from the signature data base 180 via the path 182 .
  • Each reference signature set includes an ordered series of reference signatures S REF in the order in which they were generated from the corresponding reference content.
  • the video search engine 185 determines the similarity between test content and a particular reference content by calculating a measure of dissimilarity DSM between the test signature set for the test content and the reference signature set for the particular reference content.
  • This measure of dissimilarity DSM is derived from the Hamming distances between corresponding signatures in the series of signatures for the test signature set and the reference signature set for the particular reference content. This measure may be calculated in a number of ways including either of the following expressions:
  • DSM the calculated measure of dissimilarity
  • HD[x,y] the Hamming distance between signatures x and y;
  • the video search engine 185 searches the signature data base 180 for the reference signature set that yields the smallest measure of dissimilarity with the test signature set.
  • the reference content associated with this reference signature set is the most likely candidate in the data base to share a common origin with the test content. If the measure of dissimilarity is less than some classification threshold, the test content associated with the test signature set is deemed to share a common origin with or be a copy of the reference content that is associated with the matching reference signature set.
  • Empirical results suggest that good results can be obtained for a variety of video content using if the series of signatures in each signature set represent about two seconds of video content.
  • test content and some specified reference content are said to be “matching” if the test content shares a common origin with the specified reference content.
  • the value that is chosen for the classification threshold mentioned above affects the likelihood that test and reference content will be correctly recognized as either matching or not matching each other. It also affects the likelihood that an incorrect decision is made.
  • the probability of an “incorrect negative decision” that matching content will be incorrectly classified as content that does not match increases as the value of the classification threshold decreases. Conversely, the probability of an “incorrect positive decision” that non-matching content will be incorrectly classified as content that does match increases as the value of the classification threshold increases.
  • the classification threshold may be set in any way that may be desired.
  • One method that may be used to set the value of the classification threshold obtains the original video content that is represented by a reference signature set in the data base 180 and creates a number of copies of this original content. The copies are modified in a variety of ways such as by frame-rate conversion and any of the other intentional and unintentional modifications described above.
  • the method generates a test signature set for each copy and calculates a first set of measures of dissimilarity DSM between the test signature sets and the reference signature set.
  • the method also calculates a second set of measures of dissimilarity DSM between the test signature sets and the signature sets for other video content that do not share a common origin with the original content. The range of values in the two sets may not overlap.
  • the classification threshold is set to a value within the overlap or between the two ranges if they do not overlap. This threshold value may be adjusted according to the needs of the application to balance the risk of incurring either incorrect positive or incorrect negative decisions.
  • FIG. 8 is a schematic block diagram of a device 70 that may be used to implement aspects of the present invention.
  • the processor 72 provides computing resources.
  • RAM 73 is system random access memory (RAM) used by the processor 72 for processing.
  • ROM 74 represents some form of persistent storage such as read only memory (ROM) for storing programs needed to operate the device 70 and possibly for carrying out various aspects of the present invention.
  • I/O control 75 represents interface circuitry to receive and transmit signals by way of the communication channels 76 , 77 .
  • all major system components connect to the bus 71 , which may represent more than one physical or logical bus; however, a bus architecture is not required to implement the present invention.
  • additional components may be included for interfacing to devices such as a keyboard or mouse and a display, and for controlling a storage device 78 having a storage medium such as magnetic tape or disk, or an optical medium.
  • the storage medium may be used to record programs of instructions for operating systems, utilities and applications, and may include programs that implement various aspects of the present invention.
  • Software implementations of the present invention may be conveyed by a variety of machine readable media such as baseband or modulated communication paths throughout the spectrum including from supersonic to ultraviolet frequencies, or storage media that convey information using essentially any recording technology including magnetic tape, cards or disk, optical cards or disc, and detectable markings on media including paper.
  • machine readable media such as baseband or modulated communication paths throughout the spectrum including from supersonic to ultraviolet frequencies, or storage media that convey information using essentially any recording technology including magnetic tape, cards or disk, optical cards or disc, and detectable markings on media including paper.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Library & Information Science (AREA)
  • Signal Processing (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Television Systems (AREA)
  • Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
  • Compression Or Coding Systems Of Tv Signals (AREA)
  • Studio Devices (AREA)
  • Image Processing (AREA)
US12/600,466 2007-05-17 2008-05-01 Deriving video signatures that are insensitive to picture modification and frame-rate conversion Expired - Fee Related US8488061B2 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US12/600,466 US8488061B2 (en) 2007-05-17 2008-05-01 Deriving video signatures that are insensitive to picture modification and frame-rate conversion

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
US93090507P 2007-05-17 2007-05-17
US12/600,466 US8488061B2 (en) 2007-05-17 2008-05-01 Deriving video signatures that are insensitive to picture modification and frame-rate conversion
PCT/US2008/005588 WO2008143768A1 (en) 2007-05-17 2008-05-01 Deriving video signatures that are insensitive to picture modification and frame-rate conversion

Publications (2)

Publication Number Publication Date
US20100238350A1 US20100238350A1 (en) 2010-09-23
US8488061B2 true US8488061B2 (en) 2013-07-16

Family

ID=39580355

Family Applications (1)

Application Number Title Priority Date Filing Date
US12/600,466 Expired - Fee Related US8488061B2 (en) 2007-05-17 2008-05-01 Deriving video signatures that are insensitive to picture modification and frame-rate conversion

Country Status (7)

Country Link
US (1) US8488061B2 (de)
EP (1) EP2149098B1 (de)
JP (1) JP5143896B2 (de)
CN (1) CN101681373B (de)
AT (1) ATE494587T1 (de)
DE (1) DE602008004340D1 (de)
WO (1) WO2008143768A1 (de)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190005242A1 (en) * 2017-06-28 2019-01-03 Apple Inc. Determining the Similarity of Binary Executables

Families Citing this family (30)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8351643B2 (en) * 2007-10-05 2013-01-08 Dolby Laboratories Licensing Corporation Media fingerprints that reliably correspond to media content
GB2457694B (en) 2008-02-21 2012-09-26 Snell Ltd Method of Deriving an Audio-Visual Signature
US8406462B2 (en) 2008-08-17 2013-03-26 Dolby Laboratories Licensing Corporation Signature derivation for images
US8428301B2 (en) * 2008-08-22 2013-04-23 Dolby Laboratories Licensing Corporation Content identification and quality monitoring
EP2324475A1 (de) * 2008-08-26 2011-05-25 Dolby Laboratories Licensing Corporation Robuste medien-fingerabdrücke
EP2366170B1 (de) * 2008-11-17 2013-01-02 Dolby Laboratories Licensing Corporation Medien-fingerabdrücke, die zuverlässig medieninhalt entsprechen, mit projektion von moment-invarianten
US8571255B2 (en) 2009-01-07 2013-10-29 Dolby Laboratories Licensing Corporation Scalable media fingerprint extraction
US9075897B2 (en) 2009-05-08 2015-07-07 Dolby Laboratories Licensing Corporation Storing and searching fingerprints derived from media content based on a classification of the media content
US9071868B2 (en) 2009-05-29 2015-06-30 Cognitive Networks, Inc. Systems and methods for improving server and client performance in fingerprint ACR systems
US9449090B2 (en) 2009-05-29 2016-09-20 Vizio Inscape Technologies, Llc Systems and methods for addressing a media database using distance associative hashing
US10949458B2 (en) 2009-05-29 2021-03-16 Inscape Data, Inc. System and method for improving work load management in ACR television monitoring system
US8769584B2 (en) 2009-05-29 2014-07-01 TVI Interactive Systems, Inc. Methods for displaying contextually targeted content on a connected television
US10116972B2 (en) 2009-05-29 2018-10-30 Inscape Data, Inc. Methods for identifying video segments and displaying option to view from an alternative source and/or on an alternative device
WO2010144671A2 (en) 2009-06-11 2010-12-16 Dolby Laboratories Licensing Corporation Trend analysis in content identification based on fingerprinting
PL2317517T3 (pl) * 2009-10-09 2014-09-30 Adelphoi Ltd Generowanie zapisu metadanych
US8521779B2 (en) 2009-10-09 2013-08-27 Adelphoi Limited Metadata record generation
US8786785B2 (en) * 2011-04-05 2014-07-22 Microsoft Corporation Video signature
CN105144141B (zh) * 2013-03-15 2018-12-07 构造数据有限责任公司 用于使用距离关联性散列法对媒体数据库定址的系统和方法
US9955192B2 (en) 2013-12-23 2018-04-24 Inscape Data, Inc. Monitoring individual viewing of television events using tracking pixels and cookies
GB2527528B (en) 2014-06-24 2021-09-29 Grass Valley Ltd Hash-based media search
BR112017016123A2 (pt) 2015-01-30 2018-04-17 Inscape Data Inc servidor de correspondência para identificação de conteúdo de vídeo que é exibido por um sistema de televisão, método executado por computador, e produto de programa informático concretamente incorporado a um meio de armazenamento de leitura por máquina permanente de um dispositivo de informática
US9846919B2 (en) 2015-02-16 2017-12-19 Samsung Electronics Co., Ltd. Data processing device for processing multiple sensor data and system including the same
US10410398B2 (en) * 2015-02-20 2019-09-10 Qualcomm Incorporated Systems and methods for reducing memory bandwidth using low quality tiles
WO2016151415A1 (en) * 2015-03-25 2016-09-29 Cisco Technology, Inc. Storing and retrieval heuristics
EP4375952A3 (de) 2015-04-17 2024-06-19 Inscape Data, Inc. Systeme und verfahren zur verringerung der datendichte in grossen datensätzen
US10080062B2 (en) 2015-07-16 2018-09-18 Inscape Data, Inc. Optimizing media fingerprint retention to improve system resource utilization
AU2016291674B2 (en) 2015-07-16 2021-08-26 Inscape Data, Inc. Systems and methods for partitioning search indexes for improved efficiency in identifying media segments
KR102583180B1 (ko) 2015-07-16 2023-09-25 인스케이프 데이터, 인코포레이티드 공통적인 미디어 세그먼트의 검출
KR102690528B1 (ko) 2017-04-06 2024-07-30 인스케이프 데이터, 인코포레이티드 미디어 시청 데이터를 사용하여 디바이스 맵의 정확도를 향상시키는 시스템 및 방법
US9936230B1 (en) 2017-05-10 2018-04-03 Google Llc Methods, systems, and media for transforming fingerprints to detect unauthorized media content items

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5019899A (en) * 1988-11-01 1991-05-28 Control Data Corporation Electronic data encoding and recognition system
US5465353A (en) 1994-04-01 1995-11-07 Ricoh Company, Ltd. Image matching and retrieval by multi-access redundant hashing
US5870754A (en) 1996-04-25 1999-02-09 Philips Electronics North America Corporation Video retrieval of MPEG compressed sequences using DC and motion signatures
US20040240562A1 (en) 2003-05-28 2004-12-02 Microsoft Corporation Process and system for identifying a position in video using content-based video timelines
US20050018925A1 (en) * 2003-05-29 2005-01-27 Vijayakumar Bhagavatula Reduced complexity correlation filters
US20050175224A1 (en) 2004-02-11 2005-08-11 Microsoft Corporation Desynchronized fingerprinting method and system for digital multimedia data
US20060184963A1 (en) * 2003-01-06 2006-08-17 Koninklijke Philips Electronics N.V. Method and apparatus for similar video content hopping

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2003169309A (ja) * 2001-12-04 2003-06-13 Nippon Hoso Kyokai <Nhk> コンテンツ特徴量抽出装置及びそのプログラム並びにコンテンツ認証データ生成装置及びそのプログラム並びにコンテンツ認証方法
US7532804B2 (en) * 2003-06-23 2009-05-12 Seiko Epson Corporation Method and apparatus for video copy detection

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5019899A (en) * 1988-11-01 1991-05-28 Control Data Corporation Electronic data encoding and recognition system
US5465353A (en) 1994-04-01 1995-11-07 Ricoh Company, Ltd. Image matching and retrieval by multi-access redundant hashing
US5870754A (en) 1996-04-25 1999-02-09 Philips Electronics North America Corporation Video retrieval of MPEG compressed sequences using DC and motion signatures
US20060184963A1 (en) * 2003-01-06 2006-08-17 Koninklijke Philips Electronics N.V. Method and apparatus for similar video content hopping
US20040240562A1 (en) 2003-05-28 2004-12-02 Microsoft Corporation Process and system for identifying a position in video using content-based video timelines
US20050018925A1 (en) * 2003-05-29 2005-01-27 Vijayakumar Bhagavatula Reduced complexity correlation filters
US20050175224A1 (en) 2004-02-11 2005-08-11 Microsoft Corporation Desynchronized fingerprinting method and system for digital multimedia data

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
J. Fridrich and M. Goljan "Robust Hash Functions for Digital Watermarking" Proceedings International Conference on Information Technology: Coding and Computing, 2000. *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190005242A1 (en) * 2017-06-28 2019-01-03 Apple Inc. Determining the Similarity of Binary Executables
US10685113B2 (en) * 2017-06-28 2020-06-16 Apple Inc. Determining the similarity of binary executables

Also Published As

Publication number Publication date
US20100238350A1 (en) 2010-09-23
EP2149098A1 (de) 2010-02-03
ATE494587T1 (de) 2011-01-15
JP5143896B2 (ja) 2013-02-13
EP2149098B1 (de) 2011-01-05
CN101681373A (zh) 2010-03-24
JP2010527556A (ja) 2010-08-12
CN101681373B (zh) 2012-09-26
WO2008143768A1 (en) 2008-11-27
DE602008004340D1 (de) 2011-02-17

Similar Documents

Publication Publication Date Title
US8488061B2 (en) Deriving video signatures that are insensitive to picture modification and frame-rate conversion
US8351643B2 (en) Media fingerprints that reliably correspond to media content
US8259806B2 (en) Extracting features of video and audio signal content to provide reliable identification of the signals
CN102292726B (zh) 视频标识符提取设备
Yerushalmy et al. Digital image forgery detection based on lens and sensor aberration
US11417076B2 (en) Detecting a sub-image region of interest in an image using pilot signals
US9679366B2 (en) Guided color grading for extended dynamic range
US6771795B1 (en) Spatio-temporal channel for image watermarks or data
CN113330499B (zh) 传感器装置和加密方法
KR20140058643A (ko) 강건한 낮은 복잡도 비디오 핑거프린팅을 위한 장치 및 방법
Kumar et al. A review of different prediction methods for reversible data hiding
Mehta et al. Near-duplicate detection for LCD screen acquired images using edge histogram descriptor
EP1654703B1 (de) Detektion von grafik-überlagerungen
Muratov et al. Saliency detection as a support for image forensics
CN110619362B (zh) 一种基于感知与像差的视频内容比对方法及装置
RoselinKiruba et al. Hiding data in videos using optimal selection of key-frames
Warbhe et al. Survey on pixel and format based image forgery detection techniques
KR101601755B1 (ko) 영상 특징 추출 방법 및 장치 및 이를 구현한 프로그램을 기록한 기록 매체
Yang et al. Universal steganalysis to images from normal and forged images
JP2000101834A (ja) 入力情報補正方法及びそのプログラムを記録した記録媒体

Legal Events

Date Code Title Description
REMI Maintenance fee reminder mailed
LAPS Lapse for failure to pay maintenance fees
STCH Information on status: patent discontinuation

Free format text: PATENT EXPIRED DUE TO NONPAYMENT OF MAINTENANCE FEES UNDER 37 CFR 1.362

FP Lapsed due to failure to pay maintenance fee

Effective date: 20170716